,3,,. AN EVOLUTIONIST IN EAST AFRICA. 747 



refused by fowls. I offered the geckos some "fireflies" (a small 

 luminous beetle, not an Elater) which, though examined, were not 

 eaten, so far as I saw. At Mombasa I offered my geckos two moths 

 closely resembling the Buff Ermine, and presumably of the same 

 genus. One was seized, and it did not appear to be rejected. A 

 glossy-black Blaps-like beetle was not touched either by chamaeleon 

 or gecko, nor was the moth Egyholia vaillantina, a specimen of whicli 

 I knocked down and gave to them. This moth is dark metallic 

 blue, marked slightly with yellow, and looks black and very con- 

 spicuous in its slow heavy flight. The chamaeleons on this occasion 

 were probably not feeding, and I shortly afterwards found the 

 moth dead — possibly injured in its capture. Wood-lice also were not 

 touched by my small lizards, though Gerrhosaurus major, of which I 

 had living specimens, readily ate a large Armadillo. Yet I once saw 

 one chewed and rejected by one of these large lizards. Both geckos 

 and a Mahuia seized and refused a red, black-legged Hemipteron 

 common at Mombasa. The large milliped Spirobolus, already men- 

 tioned in Natural Science, is common near Mombasa, and Dr. 

 Baxter of the Mission informed me that the odour it emitted when 

 attacked by a mongoose was so pungent as not only to baffle this- 

 enemy, but to make his eyes smart as he stood by. A specimen 

 procured at Wasin was offered to my own mongoose, which refused' 

 to attack it. As this milliped is dead black, with red legs and 

 antennae, it would appear to be a good instance of warning coloration. 

 On offering the large dark-brown species, Spirosti'eptiis gigas, to a 

 Galago which readily ate grasshoppers, however, it was untouched.- 

 Possibly, therefore, this also is unpalatable, though not marked with 

 warning colours. 



I regret that I have not been able to make any noteworthy ob- 

 servations bearing on sexual selection. I once, however, was witness 

 to what Mr. Boulenger thinks was the courtship of two lacertilians, 

 though I interrupted it, under the impression that it was a 

 fight, by attempting to catch them as they circled round and round 

 with their heads down and tails up, occasionally closing, biting, and 

 rolling over. The species was Mahuia striata, which, though hand- 

 some above, is quite plain beneath. One is at first very much tempted 

 to put down to sexual coloration the red patches at the back of the thigh 

 which are found in the toad, Bufo regularis, in this locality, and about 

 the same region and vmder the arms in a tree-frog, Hylamhates maailaius 

 As these tints are concealed when the animal is in its normal squatting" 

 position, one naturally thinks of some special reason for their exis- 

 tence. Mr. Boulenger considers that they cannot be sexual, as they 

 exist in both sexes, and as the pairing in the Anura is not calculated 

 for the display of beauty on the part of the animals. The tendency 

 to red marking in widely-separated genera of Anura in certain local 

 areas is also exemplified in the Malay Archipelago' ; in a collectioa 



^ See Boulenger, P.Z.S., 1890, pp. 31-40. 



